[R] SARIMA simulation using time series history

[Knut Erik Vedahl]

Thank you for your suggestion, Juan. However, I have tried to specify  
the innovations and this doesn't seem to solve the problem.

Are there really no opportunities to simulate from a (S)ARIMA model  
using past observations in R, besides to make the whole script from  
the ground?


Knut


Siterer Juan Pablo Calle <mi.rhelp.es at gmail.com>:

> Hi Knut,
>
> The arima.sim function, in the stats package, has an argument  
> innov="YourSerie" for the innovations, you should try something like:
>
> arima.sim(list(order=c(p,d,q), ar=c(ar coefficients), ma=c(ma  
> coefficients), seasonal=list(order=c(P,D,Q), period=NA)), n=N,  
> innov=YourSerie)
>
> I'm not sure, but maybe this help you.
>
> Salute.
>
> Juan.
>
> On 2010-10-28, at 8:08 PM, Knut Erik Vedahl wrote:
>
>> Hi,
>>
>> I'm currently working with a SARIMA model from which I want to make  
>> simulations. As I understand, neither sarima.Sim nor the functions  
>> in the gsarima package use historic realizations of the time series  
>> to simulate future values. However, I want to use historic values  
>> as input and simulate future values based on the history. Anyone  
>> who know whether such a function is available in the libaries or  
>> anywhere else?
>>
>>
>> Thanks for any help!
>>
>> Knut
>>
>> ______________________________________________
>> R-help at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>
>
>